Identification of Rare Cortical Folding Patterns using Unsupervised Deep Learning
This work addresses the challenge of detecting rare brain folding patterns for potential use in diagnosing neurodevelopmental disorders, representing an incremental improvement in domain-specific applications.
The paper tackled the problem of identifying rare cortical folding patterns in brain MR images, which are potential biomarkers for neurodevelopmental disorders, by proposing an unsupervised deep learning approach using a beta-VAE; the results showed that the method generalizes well to different regions and datasets, enabling the identification of rare patterns with complementary information from latent space and reconstruction errors.
Like fingerprints, cortical folding patterns are unique to each brain even though they follow a general species-specific organization. Some folding patterns have been linked with neurodevelopmental disorders. However, due to the high inter-individual variability, the identification of rare folding patterns that could become biomarkers remains a very complex task. This paper proposes a novel unsupervised deep learning approach to identify rare folding patterns and assess the degree of deviations that can be detected. To this end, we preprocess the brain MR images to focus the learning on the folding morphology and train a beta-VAE to model the inter-individual variability of the folding. We compare the detection power of the latent space and of the reconstruction errors, using synthetic benchmarks and one actual rare configuration related to the central sulcus. Finally, we assess the generalization of our method on a developmental anomaly located in another region. Our results suggest that this method enables encoding relevant folding characteristics that can be enlightened and better interpreted based on the generative power of the beta-VAE. The latent space and the reconstruction errors bring complementary information and enable the identification of rare patterns of different nature. This method generalizes well to a different region on another dataset. Code is available at https://github.com/neurospin-projects/2022_lguillon_rare_folding_detection.